From Knowledge Discovery
July 23, 2008 (Wednesday)
| Welcome to Penn Data Mining Group!
|
|
|
| Overview
|
| The Penn Data Mining Group develops principled means of modeling and analyzing large data sets. We combine modern statistical methods, machine learning, and knowledge of specific application areas to develop new approaches to data mining.
|
| Current projects include developing new methods of
|
- clustering and analyzing hundreds of thousands of articles using citations as well as the text
- selecting the best of hundreds of thousands of variables to use in a model
- statistical learning from relational data
- learning gene regulatory networks in brain, circadian rhythm and liver.
|
| See our Research Descriptions and slightly out of date Publications for more info. We also have a Data Mining Reading Group with a mailing list, data-mining-group@lists.seas.upenn.edu, which you can subscribe to.
|
| Computer Science Contact
|
| Lyle Ungar
|
| ungar@cis.upenn.edu
|
| Statistics Department Contact
|
| 'Dean Foster'
|
| dean@foster.net
|
|
|
|
|
